1. Field of the Invention
This invention relates to a distortion compensator that compensates for distortion arising in an amplifier that amplifies signals, and particularly to a distortion compensator that, in a configuration whereby distortion compensation is performed based on the correspondence between signal levels and distortion compensation control values that determine the mode of distortion compensation, is intended to improve the efficiency of distortion compensation by controlling the number of pairs of corresponding signal levels and distortion compensation control values, controlling the amount by which the correspondence between signal levels and distortion compensation control values is updated, and controlling the frequency at which the correspondence between signal levels and distortion compensation control values is updated.
2. Description of the Prior Art
In a base station unit provided in a mobile communications system that adopts wideband-code division multiple access (W-CDMA) as the mobile communications protocol, it is necessary that the wireless signals reach multiple mobile station units that are physically far apart, so amplifiers must be used to greatly amplify the signals to be transmitted before transmission.
However, amplifiers are analog devices, so their input/output characteristics are nonlinear functions. In particular, after the amplification limit called the saturation point, the output power becomes nearly constant even if the power input to the amplifier is increased further, so this nonlinear output gives rise to nonlinear distortion.
In the transmit signal prior to amplification, signal components outside of the desired signal band are kept to low levels by bandlimiting filters, but in the signal after passing through the filter, nonlinear distortion occurs so signal components may leak to adjacent channels or other areas outside the desired signal band. For example, in a base station unit, the transmission power is large as described above, so the magnitude of this leakage of power to adjacent channels is strictly regulated, so techniques of decreasing this adjacent channel leak power (ACP) are used.
An example of a technique of decreasing the adjacent channel leak power is the use of a predistorter.
Based on a distortion compensation table containing input signal levels and the corresponding distortion compensation control values, a predistorter generates distortion that will cancel the distortion generated in the amplifier with respect to the signal amplified by the amplifier, thus compensating for the distortion generated in the amplifier and decreasing the adjacent channel leak power.
In recent years, the adaptive predistortion (APD) method in which the contents of the distortion compensation table are adaptively controlled is attracting attention as an example of a distortion compensation scheme for implementing a highly efficient amplifier, so adaptive predistorters that perform such control are attracting attention.
Here follows a description of background art on distortion compensation.
With the “Linearizer” recited in the present inventors' publication of unexamined Japanese patent application (Kokai) No. JP-A-2001-368150 (Reference Document 1), the scope of variation of the input signal to the power amplifier is divided into a plurality of divisions, the points indicating the input signal level of each division are taken to be representative points, the inverse characteristic of the nonlinearity of the power amplifier is calculated for only the representative points and a coefficient of distortion compensation is found, and then the coefficient of distortion compensation for the representative points is used for interpolation and inverse interpolation to find a coefficient of distortion compensation for other points of the input signal level. In addition, this Reference Document 1 also recites interpolation performed using Lagrange interpolation polynomials, improving the accuracy of interpolation by increasing the order of such interpolation polynomials and also improving the accuracy of interpolation by making smaller the intervals between the representative points of the input signal level.
With the “Predistortion-Type Nonlinear Distortion Compensation Circuit and Digital Transmitter Using Same” recited in Kokai No. JP-A-2001-284980 (Reference Document 2), interpolation or extrapolation are performed using proportional calculation, and when the distortion compensation value table is updated a stipulated number of times, distortion compensation values that had never been updated are updated by interpolation. Specifically, assuming that p1<p2, if (p1, q1) and (p2, q2) are known then interpolation to calculate (p, q) with respect to p such that p1<p<p2, and extrapolation to calculate (p, q) with respect to p such that p<p1<p2 and p1<p2<p are performed.
With the “Signal Distortion Compensator and Distortion Compensation Method” recited in Kokai No. JP-A-2002-111401 (Reference Document 3), the coefficient of distortion compensation is updated based on the input signal and signal amplified by the amplifier. Specifically, the least mean squares (LMS) algorithm and the clipped least mean squares algorithm are used to update the coefficient of distortion compensation. In addition, the value of the step size of the clipped least mean squares algorithm is controlled and the dynamic range of the analog to digital (A/D) converter is controlled.
With the “Nonlinear Distortion-Compensated Power Amplifier” recited in Kokai No. JP-A-2001-203539 (Reference Document 4), the content of the table containing the input signal power values to the power amplifier and the corresponding distortion compensation coefficients (control coefficient values) is updated.
With the “Distortion Compensation Circuit” recited in Kokai No. JP-A-11-136302 (Reference Document 5), the gain deviation and orthogonal error of the quadrature modulator that causes deterioration of the accuracy of distortion compensation are compensated.
However, with a conventional predistorter, although the content of the distortion compensation table is updated, further improvement in efficiency is desirable. Specifically, with a conventional predistorter, when the content of the distortion compensation table is updated, it is desirable to improve the speed of convergence of the distortion compensation table, improve the accuracy of the content of the distortion compensation table and achieve lower power consumption.
The present invention came about in light of these conventional circumstances and has as its object to provide a distortion compensator that is able to improve the efficiency of distortion compensation at the time of compensating for the distortion arising in an amplifier based on the correspondence between signal levels and distortion compensation control values that determines the mode of distortion compensation. Specifically, with the present invention, when the correspondence between signal levels and distortion compensation control values is updated, it is possible to improve the speed of convergence of the distortion compensation table, improve the accuracy of the content of the distortion compensation table and achieve lower power consumption.
Note that in comparison to the present invention to be described later, with the background art presented in Reference Documents 1–5 above, there is no recitation or suggestion of adaptive control of the number of distortion compensation control values which is performed in the present invention, of adaptive control of the amount by which the correspondence between signal levels and distortion compensation control values is updated which is performed in the present invention, or of adaptive control of the frequency at which the correspondence between signal levels and distortion compensation control values is updated.